agentleFS
nuncaeslupusnuncaeslupus/literary-critic/.claude/skills/pin-check/SKILL.md

pin-check

Skill0 starsChanged 6 days agoMITView on GitHub

Reverts a line, runs the scoped test and reports whether it pins the behaviour. Use when the user asks if a test catches a change. Not for mutation scoring.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the script rather than a hand sed cycle: it purges __pycache__, runs each test in a fresh subprocess, refuses a target that matches …

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

pin-check

One deliberate mutation per claim, aimed at one test: is this behaviour pinned, or merely true? coverage-gaps answers whether a line ran and mutmut-report scores thousands of automatic mutants; this answers one claim in seconds.

CANARY: pin-check-loaded-2026-09-15-1c4de0b7-9a2f61d08e4b7c35

Step 1 — Name the claim and its one test

Give the exact text the claim is about and the one test that should catch it changing. Scoping to that test keeps a run to seconds.

python3 "${CLAUDE_SKILL_DIR}/scripts/pin_check.py" \
    --source src/yourproject/rules.yaml \
    --replace "asynchronous" --with "async" \
    --test tests/<the-one-case>.py

--expect-count N refuses to run unless the target occurs exactly N times, so one claim stays one mutation. --root DIR (repeatable) names extra trees to purge bytecode from. Arguments after the flags go to pytest.

Use the script rather than a hand sed cycle: it purges __pycache__, runs each test in a fresh subprocess, refuses a target that matches nothing, and restores the file when the run ends; after a killed run, its sentinel makes the next run refuse to start and print the manual restore command. A hand cycle gets each of these wrong without noticing.

Step 2 — Read the verdict

verdictexitmeans
PINNED0the target changed and the scoped test went red; the claim holds
NOT PINNED1the test stayed green; it asserts something true, not something pinned
NOT MUTATED3the target was absent or matched a different count than --expect-count; nothing was tested
INCONCLUSIVE4the test was already red before the mutation, or the file could not be restored
error2bad arguments, or a sentinel from a killed run is on disk (the message says how to restore)

Step 3 — Act on it

  • NOT PINNED → add an assertion on the value that changed, or a case that exercises the path, then re-run until PINNED.
  • PINNED → cite it in the review.
  • Either way, report the mutation — source, target, replacement, test — with the verdict, since the verdict is only as good as the mutation.

Files outside version control, or generated files, cannot be safely restored; check them in first.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.